|
on Discrete Choice Models |
| By: | Massimiliano Bratti (DEAS, University of Milan) |
| Abstract: | Although past research has established the existence of strong social class effects on the decision to undertake higher education in the UK, there is only sparse empirical work investigating social class influences on the choice of degree subject at the undergraduate level. We estimate trinomial probit models of undergraduate degree subject enrolled for the period 1981-1991 using Universities' Statistical Record data and generally find no social class effect. This finding is robust to different ways to aggregate degree subjects and the use of alternative econometric models. Our analysis suggests that in a period pre-dating the mass expansion of higher education, the replacement of student grants with student loans and the introduction of undergraduate student tuition fees, the UK university system granted equal opportunities to students from different social classes in terms of the degree subject enrolled. |
| Keywords: | degree subject, social class, UK, undergraduate, |
| Date: | 2005–11–24 |
| URL: | http://d.repec.org/n?u=RePEc:bep:unimip:1015&r=dcm |
| By: | Ron Mittelhammer (Washington State University); George Judge (University of California, Berkeley and Giannini Foundation); Douglas Miller (Purdue University); N. Scott Cardell (Salford Systems, Inc., San Diego, CA) |
| Abstract: | This paper introduces a new class of estimators based on minimization of the Cressie-Read (CR)power divergence measure for binary choice models, where neither a parameterized distribution nor a parameterization of the mean is specified explicitly in the statistical model. By incorporating sample information in the form of conditional moment conditions and estimating choice probabilities by optimizing a member of the set of divergence measures in the CR family, a new class of nonparametric estimators evolves that requires less a priori model structure than conventional parametric estimators such as probit or logit. Asymptotic properties are derived under general regularity conditions and finite sampling properties are illustrated by Monte Carlo sampling experiments. Except for some special cases in which the general regularity conditions do not hold, the estimators have asymptotic normal distributions, similar to conventional parametric estimators of the binary choice model. The sampling experiments focus on the mean square errors in the choice probability predictions and the probability derivatives with respect to the response variable values. The simulation results suggest that estimators within the CR class are more robust than conventional methods of estimation across varying probability distributions underlying the Bernoulli process. The size and power of test statistics based on the asymptotics of the CR-based estimators exhibit behavior similar to those based on conventional parametric methods. Overall, the new class of nonparametric estimators for the binary response model is a promising and potentially more robust alternative to the arametric methods often used in empirical practice. |
| Keywords: | nonparametric binary response models and estimators, conditional moment equations, finite sample bias and precision, squared error loss, response variables, Cressie-Read statistic, information theoretic methods, |
| Date: | 2005–08–01 |
| URL: | http://d.repec.org/n?u=RePEc:cdl:agrebk:998&r=dcm |
| By: | Peter Haan; Arne Uhlendorff |
| Abstract: | In this paper we suggest a Stata routine for multinomial logit models with unobserved heterogeneity using maximum simulated likelihood based on Halton sequences. The purpose of this paper is twofold: First, we provide a description of the technical implementation of the estimation routine and discuss its properties. Further, we compare our estimation routine to the Stata program gllamm which solves integration using Gauss Hermite quadrature or Bayesian adaptive quadrature. For the analysis we draw on multilevel data about schooling. Our empirical findings show that the estimation techniques lead to approximately the same estimation results. The advantage of simulation over Gauss Hermite quadrature is a marked reduction in computational time for integrals with higher dimensions. Bayesian quadrature, however, leads to very stable results with only a few quadrature points, thus the computational advantage of Halton based simulation vanishes in our example with one and two dimensional integrals. |
| Keywords: | multinomial logit model, panel data, unobserved heterogeneity, maximum simulated likelihood, Halton sequences |
| Date: | 2006 |
| URL: | http://d.repec.org/n?u=RePEc:diw:diwwpp:dp573&r=dcm |
| By: | Danny Campbell (Queen’s University Belfast); W. George Hutchinson (Queen’s University Belfast); Riccardo Scarpa (University of Waikato and University of York) |
| Abstract: | Reported in this paper are the findings from two discrete choice experiments that were carried out to address the value of a number of farm landscape improvement measures within the Rural Environment Protection (REP) Scheme in Ireland. Image manipulation software is used to prepare photorealistic simulations representing the landscape attributes across three levels to accurately represent what is achievable within the Scheme. Using a mixed logit specification willingness to pay (WTP) distributions based on the parameter estimates obtained from the individual conditional distributions are derived. These estimates are subsequently adjusted and combined to account for baselines and levels of improvement resulting from the implementation of the REP Scheme. Individual-specific WTP estimates are thus obtained for the contribution of the Scheme to rural landscapes and are subsequently contrasted with the average cost of the Scheme across the Irish adult population. Results indicate that the Scheme contributes substantial benefits to rural landscapes. |
| Keywords: | Agri-environment, Discrete choice experiments, Individual-specific WTP, Mixed logit |
| JEL: | Q51 Q24 |
| Date: | 2006–02 |
| URL: | http://d.repec.org/n?u=RePEc:fem:femwpa:2006.26&r=dcm |
| By: | Rinaldo Brau (University of Cagliari); Davide Cao (CRENoS) |
| Abstract: | This paper studies the preferences of tourists visiting the island of Sardinia (Italy), by means of a choice modelling approach. The focus is on some specific demand-enhancing effects which should confirm the feasibility of implementing sustainable tourism policies. Multinomial logit estimations reveal the strong negative effects resulting from the congestion of tourist attractions and the major transformation of coastal environments. On the other hand, recreational services and the proximity of accommodation to the beaches also seem to be important. The computation of willingness to pay measures and choice probabilities for hypothetical destinations illustrate how this kind of approach can provide useful information in determining decision processes by policy makers and development agencies. |
| Keywords: | Tourism demand, Green preferences, Choice experiments, Stated preferences |
| JEL: | Q56 L83 C25 |
| Date: | 2006–02 |
| URL: | http://d.repec.org/n?u=RePEc:fem:femwpa:2006.33&r=dcm |
| By: | Maria Salgano (Northwestern University) |
| Abstract: | This paper investigates choice between opportunity sets. I argue that individuals may prefer to have fewer options for two reasons: First, smaller choice sets may provide information and reduce the need for the agent to contemplate the alternatives. Second, contemplation costs may be increasing in the size of the choice set, making smaller sets more desirable even when they do not provide any information to the agent. I identify which of these reasons drives individual behavior in a laboratory experiment. I find strong support for both the information and cognitive overload arguments. The effects do not disappear as participants gain experience with the task. Applications of these results include firms’ choices of product variety, as costs increase with the number of products offered, and the design of government policies, such as the Medicare Drug Discount Card Program, in which older citizens can choose among numerous cards for discounts in prescription drugs. |
| Keywords: | Choice, Opportunity Sets |
| JEL: | C9 C91 |
| Date: | 2006–02 |
| URL: | http://d.repec.org/n?u=RePEc:fem:femwpa:2006.37&r=dcm |
| By: | Lesley Chiou (Department of Economics, Occidental College) |
| Abstract: | This paper quantifies the degree of competition and spatial differentiation across retail channels by exploiting a unique dataset that describes a consumer's choice of store. I estimate a consumer's choice of retailer in the sales market for DVDs among online, mass merchant, electronics, video specialty, and music stores. Using a discrete choice model, I allow for unobserved heterogeneity in preferences for store types and disutility of travel. A consumer's traveling cost varies by income, and substitution occurs proportionately more among stores of the same type. Conditional on price and distance, the average consumer still prefers Wal-Mart over most other stores. |
| Keywords: | discrete choice, retail, spatial differentiation |
| JEL: | C25 L81 |
| Date: | 2005–05 |
| URL: | http://d.repec.org/n?u=RePEc:occ:wpaper:3&r=dcm |
| By: | Lesley Chiou (Department of Economics, Occidental College) |
| Abstract: | In the movie industry, an intriguing question is why studios cluster their big theatrical hits during the Memorial Day or July 4th weekends in the early summer as opposed to the fall. This paper examines the home video industry to provide more evidence on whether booms in theatrical revenues are supply- or demand-driven. First, I find no evidence of segmentation within the home video market by genre or newness of videos. Secondly, my estimates of the seasonality within the home video market suggest that Memorial Day and July 4th may be more favorable for a theatrical release than Labor Day. |
| Keywords: | home video, seasonality, discrete choice |
| JEL: | C13 L82 |
| Date: | 2005–11 |
| URL: | http://d.repec.org/n?u=RePEc:occ:wpaper:4&r=dcm |
| By: | Lesley Chiou (Department of Economics, Occidental College); Joan Walker (Center for Transportation Studies, Boston University) |
| Abstract: | We present examples based on actual and synthetic datasets to illustrate how simulation methods can mask identification problems in the estimation of mixed logit models. Typically, simulation methods approximate an integral (without a closed form) by taking draws from the underlying distribution of the random variable of integration. Our examples reveal how a low number of draws can generate estimates that appear identified, but in fact, are either not theoretically identified by the model or not empirically identified by the data. We also show how intelligent drawing techniques require a fewer number of draws than psuedo-random to uncover identification issues. |
| Keywords: | simulation methods, discrete choice |
| JEL: | C15 C25 |
| Date: | 2005–08 |
| URL: | http://d.repec.org/n?u=RePEc:occ:wpaper:5&r=dcm |